Marr, David 19451980
Overview
Works:  31 works in 118 publications in 2 languages and 3,000 library holdings 

Roles:  Author 
Classifications:  QP475, 152.1402854 
Publication Timeline
.
Most widely held works about
David Marr
 From the retina to the neocortex : selected papers of David Marr by David Marr( Book )
 Marr's Approach to Vision by Tomaso Poggio( Book )
Most widely held works by
David Marr
Vision : a computational investigation into the human representation and processing of visual information by
David Marr(
Book
)
46 editions published between 1982 and 2010 in 3 languages and held by 1,299 WorldCat member libraries worldwide
In this work, Marr describes a general framework for understanding visual perception and touches on broader questions about how the brain and its functions can be studied and understood
46 editions published between 1982 and 2010 in 3 languages and held by 1,299 WorldCat member libraries worldwide
In this work, Marr describes a general framework for understanding visual perception and touches on broader questions about how the brain and its functions can be studied and understood
Early processing of visual information by
David Marr(
Book
)
8 editions published between 1975 and 1976 in English and Undetermined and held by 19 WorldCat member libraries worldwide
The article describes a symbolic approach to visual information proessing, and set out four principles that appear to govern the design of comples symbolic information processing systems. A computational theory of early visual information is presented, which extends to about the level of figureground separation. It includes a processoriented theory of texture vision. Most of the theory has been implemented, and examples are shown of the analysis of several natural images. This replaces AIM324 and AIM 334
8 editions published between 1975 and 1976 in English and Undetermined and held by 19 WorldCat member libraries worldwide
The article describes a symbolic approach to visual information proessing, and set out four principles that appear to govern the design of comples symbolic information processing systems. A computational theory of early visual information is presented, which extends to about the level of figureground separation. It includes a processoriented theory of texture vision. Most of the theory has been implemented, and examples are shown of the analysis of several natural images. This replaces AIM324 and AIM 334
A life by
David Marr(
Book
)
1 edition published in 1991 in English and held by 10 WorldCat member libraries worldwide
1 edition published in 1991 in English and held by 10 WorldCat member libraries worldwide
A theory of human stereo vision by
David Marr(
Book
)
3 editions published in 1977 in English and held by 7 WorldCat member libraries worldwide
An algorithm is proposed for solving the stereoscopic matching problem. The algorithm consists of five steps: (1) Each image is filtered with bar masks of four sizes that vary with eccentricity; the equivalent filters are about one octave wide. (2) Zerocrossings of the mask values are localized, and positions that correspond to terminations are found; (3) For each mask size, matching takes place between paris of zerocrossings or terminations of the same sign in the two images, for a range of disparities up to about the width of the mask's central region; (4) Wide masks can control vergence movements, thus causing small masks to come into correspondence; and (5) When a correspondence is achieved, it is written into a dynamic buffer, called the 21/2D sketch. It is shown that this proposal provides a theoretical framework for most existing psychophysical and neurophysiological data about stereopsis. Several critical experimental predictions are also made, for instance about the size of Panum's area under various conditions. The results of such experiments would tell us whether, for example, cooperativity is necessary for the fusion process
3 editions published in 1977 in English and held by 7 WorldCat member libraries worldwide
An algorithm is proposed for solving the stereoscopic matching problem. The algorithm consists of five steps: (1) Each image is filtered with bar masks of four sizes that vary with eccentricity; the equivalent filters are about one octave wide. (2) Zerocrossings of the mask values are localized, and positions that correspond to terminations are found; (3) For each mask size, matching takes place between paris of zerocrossings or terminations of the same sign in the two images, for a range of disparities up to about the width of the mask's central region; (4) Wide masks can control vergence movements, thus causing small masks to come into correspondence; and (5) When a correspondence is achieved, it is written into a dynamic buffer, called the 21/2D sketch. It is shown that this proposal provides a theoretical framework for most existing psychophysical and neurophysiological data about stereopsis. Several critical experimental predictions are also made, for instance about the size of Panum's area under various conditions. The results of such experiments would tell us whether, for example, cooperativity is necessary for the fusion process
The recognition of sharp, closely spaced edges by
David Marr(
Book
)
2 editions published in 1974 in English and held by 7 WorldCat member libraries worldwide
The recognition of sharp edges from edge and barmask convolutions with an image is studied for the special case where the separation of the edges is of the order of the masks' panelwidths. Desmearing techniques are employed to separate the items in the image. Attention is also given to parsing desmeared mask convolutions into edges and bars; to detecting edge and bar terminations; and to the detection of small blobs
2 editions published in 1974 in English and held by 7 WorldCat member libraries worldwide
The recognition of sharp edges from edge and barmask convolutions with an image is studied for the special case where the separation of the edges is of the order of the masks' panelwidths. Desmearing techniques are employed to separate the items in the image. Attention is also given to parsing desmeared mask convolutions into edges and bars; to detecting edge and bar terminations; and to the detection of small blobs
Cooperative computation of stereo disparity by
David Marr(
Book
)
4 editions published in 1976 in English and held by 6 WorldCat member libraries worldwide
The extraction of stereo disparity information from two images depends upon establishing a correspondence between them. This article analyzes the nature of the correspondence computation, and derives a cooperative algorithm that implements it. It is show that this algorithm successfully extracts information from randomdot stereograms, and its implications for the psychophysics and neurophysiology of the visual system are briefly discussed
4 editions published in 1976 in English and held by 6 WorldCat member libraries worldwide
The extraction of stereo disparity information from two images depends upon establishing a correspondence between them. This article analyzes the nature of the correspondence computation, and derives a cooperative algorithm that implements it. It is show that this algorithm successfully extracts information from randomdot stereograms, and its implications for the psychophysics and neurophysiology of the visual system are briefly discussed
Spatial disposition of axes in a generalized cylinder representation of objects that do not encompass the viewer by
David Marr(
Book
)
2 editions published in 1975 in English and held by 6 WorldCat member libraries worldwide
It is proposed that the 3D representation of an object is represented primarily by a stickfigure configuration, where each stick represents one or more axes in the object's generalized cylinder representation. The loosely hierarchical description of a stick figure is interpreted by a specialpurpose processor, able to maintain two vector and the gravitational vertical relative to a Cartesian spaceframe. It delivers information about the appearance of these vectors
2 editions published in 1975 in English and held by 6 WorldCat member libraries worldwide
It is proposed that the 3D representation of an object is represented primarily by a stickfigure configuration, where each stick represents one or more axes in the object's generalized cylinder representation. The loosely hierarchical description of a stick figure is interpreted by a specialpurpose processor, able to maintain two vector and the gravitational vertical relative to a Cartesian spaceframe. It delivers information about the appearance of these vectors
Analysis of occluding contour by
David Marr(
Book
)
4 editions published in 1976 in English and Undetermined and held by 6 WorldCat member libraries worldwide
Almost nothing can be deduced about a general 3D surface given only its occluding contours in an image, yet contour information is easily and effectively used by us to infer the shape of a surface. Therefore, implicit in the perceptual analysis of occluding contour must lie various assumptions about the viewed surfaces. The assumptions that seem most natural are (a) that the distinction between convex and concave segments reflects real properties of the viewed surface; and (b) that contiguous portions of contour arise from contiguous parts of the viewed surfacei.e., there are no invisible obscuring edges. It is proved that, for smooth surfaces, these assumptions are essentially equivalent to assuming that the viewed surface is a generalized cone. Methods are defined for finding the axis of such a cone, and for segmenting a surface constructed of several cones into its components, whose axes can then be found separately. These methods, together with the algorithms for implementing them devised by Vatan and Marr (1977), provide one link between an uninterpreted figure extracted from an image, and the 3D representation theory of Marr and Nishihara (1977). (Author)
4 editions published in 1976 in English and Undetermined and held by 6 WorldCat member libraries worldwide
Almost nothing can be deduced about a general 3D surface given only its occluding contours in an image, yet contour information is easily and effectively used by us to infer the shape of a surface. Therefore, implicit in the perceptual analysis of occluding contour must lie various assumptions about the viewed surfaces. The assumptions that seem most natural are (a) that the distinction between convex and concave segments reflects real properties of the viewed surface; and (b) that contiguous portions of contour arise from contiguous parts of the viewed surfacei.e., there are no invisible obscuring edges. It is proved that, for smooth surfaces, these assumptions are essentially equivalent to assuming that the viewed surface is a generalized cone. Methods are defined for finding the axis of such a cone, and for segmenting a surface constructed of several cones into its components, whose axes can then be found separately. These methods, together with the algorithms for implementing them devised by Vatan and Marr (1977), provide one link between an uninterpreted figure extracted from an image, and the 3D representation theory of Marr and Nishihara (1977). (Author)
The lowlevel symbolic representation of intensity changes in an image by
David Marr(
Book
)
2 editions published in 1974 in English and held by 6 WorldCat member libraries worldwide
A family of symbols is defined by which much of the useful information in an image may be represented, and its choice is justified. The family includes symbols for the various commonly occurring intensity profiles that are associated with the edges of objects, and symbols for the gradual luminance changes that provide clues about a surface's shape. It is shown that these descriptors may readily be computed from measurements similar to those made by simple cells in the visual cortex of the cat. (Author)
2 editions published in 1974 in English and held by 6 WorldCat member libraries worldwide
A family of symbols is defined by which much of the useful information in an image may be represented, and its choice is justified. The family includes symbols for the various commonly occurring intensity profiles that are associated with the edges of objects, and symbols for the gradual luminance changes that provide clues about a surface's shape. It is shown that these descriptors may readily be computed from measurements similar to those made by simple cells in the visual cortex of the cat. (Author)
On the purpose of lowlevel vision by
David Marr(
Book
)
3 editions published in 1974 in English and held by 6 WorldCat member libraries worldwide
The article advances the thesis that the purpose of lowlevel vision is to encode symbolically all of the useful information contained in an intensity array, using a vocabulary of very lowlevel symbols. Subsequent processes should have access only to this symbolic description. The reason is one of computational expediency. It allows the lowlevel processes to run almost autonomously; and it greatly simplifies the application of criteria to an image, whose representation in terms of conditions on the initial intensities, or on simple measurements made from them, is very cumbersome
3 editions published in 1974 in English and held by 6 WorldCat member libraries worldwide
The article advances the thesis that the purpose of lowlevel vision is to encode symbolically all of the useful information contained in an intensity array, using a vocabulary of very lowlevel symbols. Subsequent processes should have access only to this symbolic description. The reason is one of computational expediency. It allows the lowlevel processes to run almost autonomously; and it greatly simplifies the application of criteria to an image, whose representation in terms of conditions on the initial intensities, or on simple measurements made from them, is very cumbersome
A note on the computation of binocular disparity in a symbolic, lowlevel visual processor by
David Marr(
Book
)
2 editions published in 1974 in English and held by 6 WorldCat member libraries worldwide
The goals of the computation that extracts disparity from pairs of pictures of a scene are defined, and the constraints imposed upon that computation by the threedimensional structure of the world are shown to be inadequate. A precise expression of the goals of the computation is possible in a lowlevel symbolic visual processor: the constraints translate in this environment to prerequisites on the binding of disparity values to lowlevel symbols. The outline of a method based on this is given
2 editions published in 1974 in English and held by 6 WorldCat member libraries worldwide
The goals of the computation that extracts disparity from pairs of pictures of a scene are defined, and the constraints imposed upon that computation by the threedimensional structure of the world are shown to be inadequate. A precise expression of the goals of the computation is possible in a lowlevel symbolic visual processor: the constraints translate in this environment to prerequisites on the binding of disparity values to lowlevel symbols. The outline of a method based on this is given
Representation and recognition of the spatial organization of three dimensional shapes by
David Marr(
Book
)
1 edition published in 1976 in English and held by 6 WorldCat member libraries worldwide
A method is given for representing 3D shapes. It is based on a hierarchy of stick figures (called 3D models), where each stick corresponds to an axis in the shape's generalized cone representation. Although the representation of a complete shape may contain many stick figures at different levels of detail, only one stick figure is examined at a time while the representation is being used to interpret an image. By thus balancing scope of description against detail, the complexity of the computations needed to support the representation is minimized. The method requires (a) a database of stored stick figures; (b) a simple device called the imagespace processor for moving between objectcentered and viewercentered coordinate frames; and (c) a process for relaxing a stored model onto the image during recognition. The relation of the theory to mental rotation phenomena is discussed, and some critical experimental predictions are made. (Author)
1 edition published in 1976 in English and held by 6 WorldCat member libraries worldwide
A method is given for representing 3D shapes. It is based on a hierarchy of stick figures (called 3D models), where each stick corresponds to an axis in the shape's generalized cone representation. Although the representation of a complete shape may contain many stick figures at different levels of detail, only one stick figure is examined at a time while the representation is being used to interpret an image. By thus balancing scope of description against detail, the complexity of the computations needed to support the representation is minimized. The method requires (a) a database of stored stick figures; (b) a simple device called the imagespace processor for moving between objectcentered and viewercentered coordinate frames; and (c) a process for relaxing a stored model onto the image during recognition. The relation of the theory to mental rotation phenomena is discussed, and some critical experimental predictions are made. (Author)
Theory of edge detection by
Ellen Catherine Hildreth(
Book
)
1 edition published in 1979 in English and held by 6 WorldCat member libraries worldwide
A theory of edge detection is presented. (1) Intensity changes, which occur in a natural image over a wide range of scales, are detected separately at different scales. At a given scale, this is best done by finding the zerocrossings of gradientsquared G(x, y) * (I(x, y) for image I, where G(x, y) is a twodimensional gaussian distribution, and gradientsquared is the Laplacian. (2) The physical phenomena that give rise to the intensity changes are localized. This allows one to construct rules for combining information from the different scales into a primitive description of the image. A physiological model for zerocrossing detection is proposed. (Author)
1 edition published in 1979 in English and held by 6 WorldCat member libraries worldwide
A theory of edge detection is presented. (1) Intensity changes, which occur in a natural image over a wide range of scales, are detected separately at different scales. At a given scale, this is best done by finding the zerocrossings of gradientsquared G(x, y) * (I(x, y) for image I, where G(x, y) is a twodimensional gaussian distribution, and gradientsquared is the Laplacian. (2) The physical phenomena that give rise to the intensity changes are localized. This allows one to construct rules for combining information from the different scales into a primitive description of the image. A physiological model for zerocrossing detection is proposed. (Author)
Analyzing natural images : a computational theory of texture vision by
David Marr(
Book
)
4 editions published in 1975 in English and Undetermined and held by 5 WorldCat member libraries worldwide
A theory of early and intermediate visual information processing is given, which extends to about the level of figureground separation. Its core is a computational theory of texture vision. Evidence obtained from perceptual and from computational experiments is adduced in its support. A consequence of the theory is that highlevel knowledge about the world influences visual processing later and in a different way from that currently practiced in machine vision
4 editions published in 1975 in English and Undetermined and held by 5 WorldCat member libraries worldwide
A theory of early and intermediate visual information processing is given, which extends to about the level of figureground separation. Its core is a computational theory of texture vision. Evidence obtained from perceptual and from computational experiments is adduced in its support. A consequence of the theory is that highlevel knowledge about the world influences visual processing later and in a different way from that currently practiced in machine vision
From understanding computation to understanding neural circuitry by
David Marr(
Book
)
1 edition published in 1976 in English and held by 5 WorldCat member libraries worldwide
The CNS needs to be understood at four nearly independent levels of description: (1) that at which the nature of a computation is expressed; (2) that at which the algorithms that implement a computation are characterized; (3) that at which an algorithm is committed to particular mechanisms; and (4) that at which the mechanisms are realized in hardware. In general, the nature of a computation is determined by the problem to be solved, the mechanisms that are used depend upon the available hardware, and the particular algorithms chosen depend on the problem and on the available mechanisms. Examples are given of theories at each level. (Author)
1 edition published in 1976 in English and held by 5 WorldCat member libraries worldwide
The CNS needs to be understood at four nearly independent levels of description: (1) that at which the nature of a computation is expressed; (2) that at which the algorithms that implement a computation are characterized; (3) that at which an algorithm is committed to particular mechanisms; and (4) that at which the mechanisms are realized in hardware. In general, the nature of a computation is determined by the problem to be solved, the mechanisms that are used depend upon the available hardware, and the particular algorithms chosen depend on the problem and on the available mechanisms. Examples are given of theories at each level. (Author)
Some comments on a recent theory of stereopsis by
David Marr(
Book
)
2 editions published in 1980 in English and held by 5 WorldCat member libraries worldwide
2 editions published in 1980 in English and held by 5 WorldCat member libraries worldwide
An essay on the primate retina by
David Marr(
Book
)
3 editions published in 1974 in English and held by 5 WorldCat member libraries worldwide
A review of the known physiology, anatomy, histology, and psychophysics of the retina is given. Land's retinex theory of colour vision is summarized. It is proposed that one function of the retina is to compute lightness using a twodimensional parallel algorithm. There are three stages: (1) a centresurround difference operation; (2) a threshold applied to the difference signal; (3) the inverse of (1), whose output is lightness. The operation of the midget bipolar  midget ganglion channel is analysed in detail, and a functional interpretation of various retinal structures is given. Requirements of the theory are stated concerning the arrangement and connexions of cells, and the signs of the synapses, in the inner plexiform layer
3 editions published in 1974 in English and held by 5 WorldCat member libraries worldwide
A review of the known physiology, anatomy, histology, and psychophysics of the retina is given. Land's retinex theory of colour vision is summarized. It is proposed that one function of the retina is to compute lightness using a twodimensional parallel algorithm. There are three stages: (1) a centresurround difference operation; (2) a threshold applied to the difference signal; (3) the inverse of (1), whose output is lightness. The operation of the midget bipolar  midget ganglion channel is analysed in detail, and a functional interpretation of various retinal structures is given. Requirements of the theory are stated concerning the arrangement and connexions of cells, and the signs of the synapses, in the inner plexiform layer
Representation and recognition of the movements of shapes by
David Marr(
Book
)
3 editions published in 1980 in English and held by 4 WorldCat member libraries worldwide
The problems posed by the representation and recognition of the movements of 3D shapes are analyzed. A representation is proposed for the movements of shapes that lie within the scope of Marr and Nishihara's (1978) 3D model representation of static shapes. The basic problem is, how to segment a stream of movement into pieces each of which can be described separately. The representation proposed here is based upon segmenting a movement at moments when a component axis, e.g. an arm, starts to move relative to its local coordinate frame (here, the torso). So that for example walking is divided into a sequence of the stationary states between each swing of the arms and legs, and the actual motions between the stationary points (relative to the torso, not the ground). This representation is called the statemotionstate (SMS) moving shape representation, and several examples of its application are given. (Author)
3 editions published in 1980 in English and held by 4 WorldCat member libraries worldwide
The problems posed by the representation and recognition of the movements of 3D shapes are analyzed. A representation is proposed for the movements of shapes that lie within the scope of Marr and Nishihara's (1978) 3D model representation of static shapes. The basic problem is, how to segment a stream of movement into pieces each of which can be described separately. The representation proposed here is based upon segmenting a movement at moments when a component axis, e.g. an arm, starts to move relative to its local coordinate frame (here, the torso). So that for example walking is divided into a sequence of the stationary states between each swing of the arms and legs, and the actual motions between the stationary points (relative to the torso, not the ground). This representation is called the statemotionstate (SMS) moving shape representation, and several examples of its application are given. (Author)
Analysis of a cooperative stereo algorithm by
David Marr(
Book
)
2 editions published in 1977 in English and held by 4 WorldCat member libraries worldwide
Marr and Poggio (1976) recently described a cooperative algorithm that solves the correspondence problem for stereopsis. This article uses a probabilistic technique to analyze the convergence of that algorithm, and derives the conditions governing the stability of the solution state. The actual results of applying the algorithm to randomdot stereograms are compared with the probabilistic analysis. A satisfactory mathematical analysis of the asymptotic behavior of the algorithm is possible for a suitable choice of the parameter values and loading rules, and again the actual performance of the algorithm under these conditions is compared with the theoretical predictions. Finally, some problems raised by the analysis of this type of 'cooperative' algorithm are briefly discussed. (Author)
2 editions published in 1977 in English and held by 4 WorldCat member libraries worldwide
Marr and Poggio (1976) recently described a cooperative algorithm that solves the correspondence problem for stereopsis. This article uses a probabilistic technique to analyze the convergence of that algorithm, and derives the conditions governing the stability of the solution state. The actual results of applying the algorithm to randomdot stereograms are compared with the probabilistic analysis. A satisfactory mathematical analysis of the asymptotic behavior of the algorithm is possible for a suitable choice of the parameter values and loading rules, and again the actual performance of the algorithm under these conditions is compared with the theoretical predictions. Finally, some problems raised by the analysis of this type of 'cooperative' algorithm are briefly discussed. (Author)
From the retina to the neocortex : selected papers of David Marr by
David Marr(
Book
)
1 edition published in 1991 in English and held by 1 WorldCat member library worldwide
1 edition published in 1991 in English and held by 1 WorldCat member library worldwide
more
fewer
Audience Level
0 

1  
Kids  General  Special 
Related Identities
 Vaina, Lucia 1946 Editor
 Poggio, T.
 MASSACHUSETTS INST OF TECH CAMBRIDGE ARTIFICIAL INTELLIGENCE LAB
 Massachusetts Institute of Technology Artificial Intelligence Laboratory
 Hildreth, Ellen Catherine
 Poggio, Tomaso Author
 Ullman, Shimon
 Royal Society (Great Britain) Philosophical Transactions
 Nishihara, Herbert Keith
 Nishihara, H. Keith
Useful Links
Associated Subjects
Binocular vision BrainComputer simulation Computer graphics Digital filters (Mathematics) Human information processing Image intensifiers Image processing Marr, David, Motion perception (Vision)Data processing Neural networks (Neurobiology) Optical data processing Optical pattern recognition Pattern recognition systems Retina Vision VisionData processing VisionMathematical models Visual perception Visual perceptionData processing